Multi-modal models require aligned, shared embedding spaces. However, common CLIP-based approaches need large amounts of samples and do not natively support 3D or tabular data, both of which are crucial in the medical domain. To address these issues, we revisit CLIP-style alignment by training a domain-specific 3D foundation model as an image encoder and demonstrate that modality alignment is feasible with only 62 MRI scans. Our approach is enabled by a simple embedding accumulation strategy required for training in 3D, which scales the amount of negative pairs across batches in order to stabilize training. We perform a thorough evaluation of various design choices, including the choice of backbone and loss functions, and evaluate the proposed methodology on zero-shot classification and image-retrieval tasks. While zero-shot image-retrieval remains challenging, zero-shot classification results demonstrate that the proposed approach can meaningfully align the representations of 3D MRI with tabular data.

Medical image analysis using deep learning frameworks has advanced healthcare by automating complex tasks, but many existing frameworks lack flexibility, modularity, and user-friendliness. To address these challenges, we introduce Yucca, an open-source AI framework available at https://github.com/Sllambias/yucca, designed specifically for medical imaging applications and built on PyTorch and PyTorch Lightning. Yucca features a three-tiered architecture: Functional, Modules, and Pipeline, providing a comprehensive and customizable solution. Evaluated across diverse tasks such as cerebral microbleeds detection, white matter hyperintensity segmentation, and hippocampus segmentation, Yucca achieves state-of-the-art results, demonstrating its robustness and versatility. Yucca offers a powerful, flexible, and user-friendly platform for medical image analysis, inviting community contributions to advance its capabilities and impact.

Deep learning-based models in medical imaging often struggle to generalize effectively to new scans due to data heterogeneity arising from differences in hardware, acquisition parameters, population, and artifacts. This limitation presents a significant challenge in adopting machine learning models for clinical practice. We propose an unsupervised method for robust domain adaptation in brain MRI segmentation by leveraging MRI-specific augmentation techniques. To evaluate the effectiveness of our method, we conduct extensive experiments across diverse datasets, modalities, and segmentation tasks, comparing against the state-of-the-art methods. The results show that our proposed approach achieves high accuracy, exhibits broad applicability, and showcases remarkable robustness against domain shift in various tasks, surpassing the state-of-the-art performance in the majority of cases.

Artificial intelligence is everywhere, and now a group of developers have created AI software that can tell whether you are likely to die from Covid-19 using health data. University of Copenhagen researchers fed a computer program with health data from 3,944 Danish COVID-19 patients, as well as any underlying conditions. They then trained it to look for patterns in a patients' prior illness to determine the risk factors and potential outcome from Covid-19 and found that BMI, age and being male were the highest risk factors when it came to the likelihood of dying. The results show that AI can, with up to 90 per cent certainty, determine whether an uninfected person will die of the disease if they are unlucky enough to catch it. Results from the new tool could help health officials determine who should be at the front of the line for a limited supply of vaccines, said lead author Mads Nielsen. They say this should be considered when determining who should get the vaccine first.

Experts at the University of Copenhagen, Denmark, have begun using artificial intelligence to create computer models that calculate the risk of a corona patient's needing intensive care or a ventilator. As coronavirus patients are hospitalized, it is difficult for doctors to predict which of them will require intensive care and a respirator. Many different factors come into play, some yet to be fully understood by doctors . As such, computer scientists at the University of Copenhagen are now developing computer models based on artificial intelligence that calculate the risk of an individual patient's need for a ventilator or intensive care. The new initiative is being conducted in a collaboration with Rigshospitalet and Bispebjerg Hospital.